Mechanical Design Guidelines for EMI Shielding
Shielding problems with EMI gaskets can often be traced back to improper mechanical compression.
Michael Keough and Jason Davis
Portable wireless devices are being used to connect medical devices and patient information to hospital networks. As these devices grow more sophisticated, consideration of electromagnetic interference (EMI) takes on increasing importance.
Designers of portable wireless devices employ a variety of shielding methods to combat EMI. Two of the most popular approaches are soldered enclosures and EMI gaskets paired with shield covers.
Soldered enclosures are commonplace in the current generation of wireless devices. However, the increased functionality, higher density and operating frequency, and overall complexity of next-generation wireless devices place greater demands on EMI shielding performance.
Many next-generation designs require shielding of 6–10 individual cavities on the printed circuit board (PCB). The EMI gasket and shield cover approach can work better for multicavity designs, providing the following benefits when compared with soldered enclosures:
When using the EMI gasket and shield cover approach, the electrical properties of the EMI gasket material play a significant role in shielding performance. Because EMI gasket materials typically depend on proper compression to achieve optimal electrical properties, shielding problems can often be traced back to mechanical design issues that limit compression.
This article presents the mechanical parameters of shield design that can provide good EMI protection while minimizing deformation of the shield cover and PCB.
Mechanical Design Parameters
Many mechanical design parameters can affect the level of gasket compression and of EMI shielding effectiveness. Key design parameters relate either to the shield cover and PCB design or to the gasket itself (see Table I).
|Shield Cover and PCB Design Parameters|
|E||Modulus of elasticity.|
|I||Cross-sectional moment of inertia.|
|O||Directional out-of-flatness of the surface to which gasket is to be applied (convex versus concave).|
|Ff||Total force exerted by the fasteners, snaps, screws, or other pressure points.|
|EMI Gasket Parameters|
|A||Surface area of the gasket.|
|D||Gasket softness (durometer).|
|T||Variability in gasket thickness.|
|Table I. Mechanical design parameters.|
The following sections discuss the mechanical parameters in detail and relate them to general electromechanical models that describe their end effects. Case studies illustrate how the mechanical parameters affect shielding performance.
Shield Cover and PCB Design Parameters
Modulus of Elasticity. Use shield cover materials that have a high modulus of elasticity (E) to obtain the level of compression required for sufficient EMI gasket performance. This reduces cover flexing to a minimum when the EMI gasket is under compressive load.
The formula for maximum displacement of a transversely loaded beam fixed on both sides is
This is a rough approximation of the case in which the cover bows around an EMI gasket between fastener locations.
The equation above shows that an increase in E reduces the amount of deflection in the cover. Different blends of plastics yield different elastic moduli; those with higher modulus of elasticity are better choices for shield cover material.
Moment of Inertia. Another way to increase the rigidity of the design is to increase the moment of inertia. As Equation 1 shows, the flexing or bowing phenomenon has a direct inverse relationship with the cross-sectional moment of inertia (I) of the cover. If I is doubled, the amount of flexing in the cover is effectively reduced by half. Because E and I values are indirectly proportional to deflection, an increase in either value decreases the overall deflection value. This helps to increase the compression on the gasket in the areas farthest from fastener locations, resulting in better electrical performance from the gasket. Making the enclosure deeper in the z direction and adding stiffening cross-ribs that cover long spans are simple ways to improve the moment of inertia of a shield cover.
Figure 1. Theoretical model of electrical properties of a typical EMI gasket between fastener locations
(Click to enlarge).
The shield cover's E and I values have similar effects on the electrical properties of the EMI gasket. Figure 1 shows a model of the relative electrical properties in terms of dc through-resistance between cover, gasket, and PCB ground for shield covers with both high and low E and I values. Dc resistance is commonly used in the industry to characterize EMI gaskets because it is easy to measure and is often used in general models to predict EMI gasket shielding performance at different power levels and frequencies.
Flatness of the Cover. Mechanical designers usually consider the flatness of molded parts to be an important parameter. However, they often do not consider the directionality of the flatness specification when designing a molded shield cover.
During assembly of a shield cover that bows in the convex direction (i.e., edges away from the PCB), the areas around the screw locations make contact last, bending the cover in the direction of the PCB and effectively aiding compression on the gasket in areas where screws are not present. This convex curvature positively affects the shielding properties of the gasket.
In contrast, a shield cover with concave curvature bows inward, with the edges closer to the PCB. During assembly, the fastener locations make contact first, and, because the ribs of the cover are bowed away from the PCB, the areas of the gasket between the screws compress less. Because there is no force aiding gasket compression in these areas, EMI shielding effectiveness typically degrades with distance from the fastener location.
In summary, a loose flatness specification is usually acceptable, provided it is specified as "minimum flat," "maximum convex" bow.
Fasteners. Fastener spacing is tremendously important in shield cover design. Equation 1 shows that the amount of flexing in the cover is proportional to the cube of the distance (X) between the fastener locations. If the distance between the fasteners is reduced by half, there is effectively one-eighth the amount of flexing in the cover. Less flexing means more compression on the EMI gasket, which results in better EMI shielding.
Sufficient compression of the EMI gasket is usually not an issue when screws are used as the fastener because screws can deliver an extremely high amount of closure force. However, the industry is moving toward reducing the number of screws within portable wireless devices. Many manufacturers are using molded snap fasteners for faster, more-efficient assembly.
In addition, the fasteners in many portable wireless devices are positioned only at the perimeter of the shield cover. This leaves the central area of the mechanical stack-up susceptible to flexing, which can adversely affect EMI shielding performance.
Because of these design realities, the mechanical parameters must be adjusted to maintain adequate compression on the EMI gasket while minimizing flexing of the PCB. One solution is to add a center screw to the shield cover. The screw reduces the amount of cover flexing by decreasing the distance between the external fasteners. Although adding one or more screws helps provide consistent force, it is not always possible. This is where the industry turns to gasket manufacturers for solutions.
Gasket Design Parameters
Force, Area, and Compressive Modulus. The total force (Ff) required to compress a given EMI gasket depends on the gasket's surface area (A) and softness or durometer (D).
Figure 2. Dc resistance test fixture.
(Click to enlarge)
A slight bow in the shield cover or PCB due to the reaction force from the gasket can cause degradation in shielding effectiveness. W. L. Gore & Associates Inc. (Elkton, MD) has developed a device to simulate the electrical performance of EMI gaskets as a function of the distance from the fastener locations (see Figure 2).
Test leads spaced at equal distances between fastener locations are used to measure the dc resistance of the cover/gasket between the screw locations. The dc resistance values give a reasonably good indication of how the gasket material will perform as an EMI shield interface in an actual wireless device.
When the EMI gasket is compressed, the through-the-thickness resistance in the z direction is much lower than the transverse resistance along the length of the gasket. This phenomenon has to do with the nature of the conductive particles in the gasket and the conductive path in the z direction that is created when the particles are compressed. Therefore, it is not necessary to segment the EMI gasket to test the dc resistance as a function of the distance from the screw. Figure 3 shows four gasket materials tested in the dc resistance fixture.
Next-generation materials are being developed that will minimize the flexing effect on EMI gasket performance. These materials will also alleviate some of the cover design attributes discussed previously that currently must be considered when using EMI gaskets.
Figure 3. Dc resistance versus distance of die-cut gaskets at varying durometers and form-in-place silicone elastomer gaskets.
(Click to enlarge)
Figure 3 clearly shows that softer or more-compliant gasket materials with a low durometer (yellow line) provide much more consistent conductivity from the cover to the PCB, even when there is a great distance to the nearest fastener. In general, softer gaskets do not cause as much bending of the shield cover or PCB when they are compressed to the proper height around the compression stops at the screw locations. Harder gaskets or those with a high durometer, such as the purple line in Figure 3, can cause so much bending of the shield or PCB that they can actually lose contact with the gasket between screw locations.
Thickness Variability. The thickness variability of EMI gaskets is a crucial factor for shielding performance. When thickness variability exists between screw locations (especially with covers that have low E and low I), the electrical performance of the EMI gasket can be adversely affected. The amount that dc resistance increases depends on the amount and location of the thickness variability, as well as on the properties of the gasket.
Figure 4. A die-cut gasket.
Die-cut gaskets provide the highest thickness consistency of all EMI gaskets. Figure 4 shows the outstanding thickness consistency of die-cut gaskets installed on handset covers. Other EMI gasket solutions, such as silicone-dispensed gaskets, typically have much lower thickness consistency.
Effects of Varying Mechanical Parameters
Although theoretical calculations are easy to perform using such simple geometry as rectangular beams, modern electrical devices rarely are so simple. Finite-element analysis (FEA) can be used to accurately predict the various effects that these mechanical parameters have on bowing and shielding in complex designs. FEA can predict deflections (bowing) and the resultant stresses and strains on device covers, EMI gaskets, and PCBs.
Figure 5. Displacement of high-durometer gasket with low-moment-of-inertia cover geometry.
(Click to enlarge)
For example, after using computer-aided design (CAD) software to create the geometry of the device cover, gasket, and PCB, the real-world device can be simulated by assembling it on the computer.
The designer knows how much the EMI gasket is supposed to compress at the device's fastener location, which is the height of the compression stop—the hard stop at which the screws or fasteners cease to advance the cover to the PCB.
Given the known amount of gasket compression, the FEA model can take into account the material properties and design geometry to predict with reasonable accuracy the deflections and stresses at any point in the design. Once trouble areas are identified, the mechanical parameters can be changed and the analysis rerun for robustness. Armed with this information, lighter, cheaper, and more-reliable designs can be created.
The dc resistance test fixture discussed previously and shown in Figure 2 provides a simple illustration of the modeling process. Designers used three-dimensional CAD software to model and assemble the test fixture's geometry. Before assembly, the model considered the gasket to be uncompressed. After assembly, the model considered the area around the screw holes to be flush with the PCB, compressing the gasket height from the original 0.6 mm to 0.4 mm. In theory, the entire cover (assuming no flex) should have moved 0.2 mm. The model could be analyzed to determine deflections and stresses along the length of the assembly. Although the screws were not shown in the analysis, the appropriate constraints were applied in the FEA model to simulate their clamping and restraining force.
Figure 6. Displacement of high-durometer gasket with high-moment-of-inertia cover geometry.
(Click to enlarge)
For the analysis, the model considered the shield cover material to be a polycarbonate and acrylonitrile butadiene-styrene (PC/ABS) blend, and the geometry was modified to show effects of varying moments of inertia. The PCB was assumed to be 1.57 mm thick and to be a typical FR-4 lay-up. The gasket was 30.23 mm long, 1 mm wide, and 0.6 mm thick. The gasket material properties were varied to show the effects of gasket softness on cover and PCB deflections.
Figures 5–7 are FEA outputs shown with different colors to indicate the gasket displacement in millimeters from its original, uncompressed state. During assembly, the cover material is intended to displace the gasket 0.2 mm. If the cover experiences no bending, the FEA output will show a maximum displacement of 0.2 mm and appear in red. Any color deviation from red indicates bending of the cover.
Conversely, the PCB in the area of the screw holes was fixed (no displacement) in the FEA model. Therefore, if there were no bending of the PCB, the FEA output would show a dark blue color (indicating zero displacement) after the assembly is compressed. Any deviation from the dark blue color indicates that the PCB is bending.
Figure 5 shows the output of the FEA model of a high-durometer (i.e., harder) gasket used with a PC/ABS cover that has a low-moment-of-inertia geometry. The center of this weak and flexible cover has little stiffness to adequately compress the gasket and has bowed upward about 0.18 mm, as indicated by the wide dark-blue area in the middle of the cover. The bowing leaves the gasket almost uncompressed in the middle between the fasteners, causing higher resistance values and lower EMI shielding effectiveness in the center section.
Figure 7. Displacement of softer gasket with high-moment-of-inertia cover geometry.
(Click to enlarge)
Figure 6 shows the same high-durometer gasket with a stiffening rib added to the cover to create a high-moment-of-inertia geometry. The cover has deflected only about 0.09 mm, as indicated by the wide green area in the center. However, because the gasket is still relatively hard, it has caused the PCB to flex more, as indicated by the PCB color change from blue to green under the center of the gasket. The PCB flex of about 0.06 mm could be a problem for designers. Accounting for PCB and cover flex, the net compression on the center of the gasket (~0.05 mm) is moderately greater than in the low-moment-of-inertia geometry shown in Figure 5, and therefore provides slightly better EMI shielding.
Figure 7 shows the best combination of mechanical parameters. In this case, the cover has a stiffening rib with a high moment of inertia, but the EMI gasket is softer (lower durometer). The gasket is compressed more than adequately between the fasteners (~0.1 mm), providing very good EMI shielding. The softer gasket material reduces the deflections of the cover and PCB to more acceptable levels.
EMI gasket materials depend on proper compression to achieve sufficient electrical performance. Consequently, shielding problems with EMI gaskets often are traced back to mechanical design issues that limit the amount of compression on the material.
Designing with an eye toward key mechanical parameters, such as modulus of elasticity, cross-sectional moment of inertia, the flatness of the cover, and the distance between fasteners, is essential for improved device performance and reliability. The choice of material is critical. Gaskets that have a low durometer and uniform thickness can help to offset some of the constraints associated with cover designs, providing increased robustness and improved shielding effectiveness.
McKay, Gordon. Standard Handbook for Mechanical Engineers, Beam Theory. New York: McGraw-Hill, 1967.
Reis, Brad. Handset Mechanical Design Assessment. Elkton, MD: W. L. Gore & Associates Inc., 1998.
White, Donald. A Handbook on Shielding Design Methodology and Procedures. Gainesville, VA: Interference Control Technologies, 1981.
Michael Keough and Jason Davis both work at W. L. Gore & Associates Inc. (Elkton, MD) as applications engineers. The authors can be reached at 410-506-4062.
Copyright © 2002 Medical Electronics Manufacturing